Report. Decoding of Cytoplasmic Ca 2+ Oscillations through the Spatial Signature Drives Gene Expression

Transcription

1 Current Biology 19, , May 26, 2009 ª2009 Elsevier Ltd All rights reserved DOI /j.cub Decoding of Cytoplasmic Ca 2+ Oscillations through the Spatial Signature Drives Gene Expression Report Joseph Di Capite, 1 Siaw Wei Ng, 1 and Anant B. Parekh 1, * 1 Department of Physiology, Anatomy and Genetics University of Oxford Parks Road Oxford OX1 3PT UK Summary Cytoplasmic Ca 2+ oscillations are a universal signaling mode that activates numerous cellular responses [1, 2]. Oscillations are considered the physiological mechanism of Ca 2+ signaling because they occur at low levels of stimulus intensity [3]. Ca 2+ oscillations are proposed to convey information in their amplitude and frequency, leading to activation of specific downstream targets [4 6]. Here, we report that the spatial Ca 2+ gradient within the oscillation is key. Ca 2+ oscillations in mast cells evoked over a range of agonist concentrations in the presence of external Ca 2+ were indistinguishable from those in the absence of Ca 2+ when plasmalemmal Ca 2+ extrusion was suppressed. Nevertheless, only oscillations with accompanying Ca 2+ entry through store-operated CRAC channels triggered gene expression. Increased cytoplasmic Ca 2+ buffering prevented oscillations but not gene activation. Local Ca 2+ influx and not global Ca 2+ oscillations therefore drives gene expression at physiological levels of stimulation. Rather than serving to maintain Ca 2+ oscillations by replenishing stores, we suggest that the role of oscillations might be to activate CRAC channels, thereby ensuring the generation of spatially restricted physiological Ca 2+ signals driving gene activation. Furthermore, we show that the spatial profile of a Ca 2+ oscillation provides a novel mechanism whereby a pleiotropic messenger specifically activates gene expression. Results and Discussion Since Prince and Berridge first proposed the existence of cytoplasmic Ca 2+ oscillations in blowfly salivary gland [7], these oscillations induced by receptor activation have been found in virtually all cell types [8].Ca 2+ oscillations are thought to confer several advantages over a sustained bulk Ca 2+ rise in activating Ca 2+ -dependent responses [9]. These include increased sensitivity to the Ca 2+ signal and circumvention both of the desensitization and excitotoxicity that can occur in response to a maintained Ca 2+ signal. Furthermore, different amplitudes and/or frequencies of Ca 2+ oscillation can recruit specific downstream Ca 2+ -dependent targets [4 6, 10], providing a mechanism for selective responses to the promiscuous Ca 2+ signal. The diffusion of Ca 2+ through an open Ca 2+ channel in either the plasma membrane or an intracellular organelle results in the rapid build-up of a local microdomain of elevated Ca 2+ [11]. Depending on the type of Ca 2+ channel, the Ca 2+ *Correspondence: anant.parekh@dpag.ox.ac.uk concentration within a microdomain can reach tens of micromolar, several-fold higher than the bulk cytoplasmic Ca 2+ rise [11, 12]. Ca 2+ microdomains associated with voltagegated Ca 2+ channels activate neurotransmitter release [11], open colocalized Ca 2+ -dependent K + channels [13], and trigger gene transcription [14, 15], whereas microdomains that accompany the opening of store-operated Ca 2+ releaseactivated Ca 2+ (CRAC) channels activate the Ca 2+ ATPAse pump [16], adenylyl cyclase [17], endothelial nitric oxide synthase, and Ca 2+ -dependent phospholipase A 2 [18]. Ca 2+ microdomains arising from open InsP 3 receptors in the endoplasmic reticulum stimulate juxtaposed mitochondria to generate ATP [19]. Ca 2+ oscillations arise from the opening of Ca 2+ -permeable ion channels, so we hypothesized that the subcellular spatial profile of Ca 2+ oscillations might play a role in activating selective downstream targets. If this is the case, then Ca 2+ oscillations that are identical in amplitude and frequency should differentially activate downstream targets depending on the spatial location of the underlying Ca 2+ channels. Here, we have generated Ca 2+ oscillations that have identical amplitudes and frequencies but with distinct spatial profiles and we compared their abilities to activate gene expression. To test the idea that the profile of a Ca 2+ oscillation contained important information, we generated two kinds of InsP 3 -driven Ca 2+ oscillation with distinct spatial signatures: one involved Ca 2+ release from the endoplasmic reticulum without any Ca 2+ influx, whereas the other comprised Ca 2+ release followed by Ca 2+ influx through store-operated CRAC channels. Only the latter scenario would result in significant elevation of the subplasmalemmal Ca 2+ concentration. Mast cells express cell-surface cysteinyl leukotriene type I receptors that couple to phospholipase C and which are activated by the proinflammatory molecule leukotriene C 4 (LTC 4 ) [20]. Stimulation with a submaximal concentration of LTC 4 (160 nm) in the presence of external Ca 2+ evoked a series of repetitive Ca 2+ oscillations that were maintained for at least 800 s (Figure 1A, representative of n > 500 cells with w98% of cells responding). In the absence of external Ca 2+, oscillations ran down with time such that most cells failed to generate aca 2+ signal after 400 s (Figure 1B, n > 350 cells, with virtually all cells responding). Hence Ca 2+ influx is needed to sustain the Ca 2+ oscillations. The Ca 2+ influx pathway involves the CRAC channel because (1) in whole-cell patch clamp experiments, LTC 4 activates CRAC channels via store depletion [20] and (2) block of these channels with 1 mm Gd 3+ or La 3+ results in rundown of the oscillations in the presence of external Ca 2+ and at a rate similar to that seen in the absence of external Ca 2+ (Figure S1 available online; 56 cells). Ca 2+ oscillations run down in Ca 2+ -free solution because the Ca 2+ that is released from the endoplasmic reticulum by InsP 3 is transported out of the cell, resulting in gradual depletion of the Ca 2+ store. Of the plasma membrane Ca 2+ removal mechanisms, the Ca 2+ ATPase pump is particularly effective at clearing Ca 2+ from the cytoplasm in mast cells [21]. After block of this pump with La 3+, stimulation with LTC 4 in the absence of external Ca 2+ resulted in Ca 2+ oscillations that were now sustained (Figure 1C; labeled 0 Ca 2+ +La 3+, n > 300 cells, with

2 Current Biology Vol 19 No Figure 1. Cytoplasmic Oscillations Evoked by Low Levels of LTC 4 (A) Treatment with 160 nm LTC 4 evoked Ca 2+ oscillations that were maintained in 2 mm external Ca 2+. (B) Oscillations ran down in the absence of external Ca 2+. (C) Oscillations were maintained in the absence of external Ca 2+, if the plasma membrane Ca 2+ ATPase pump was blocked with La 3+ (1 mm). (D F) Ca 2+ oscillations evoked by a lower concentration of LTC 4 (60 nm) showed a similar pattern. Each graph depicts different cells in a field of view. Different colors correspond to different cells. In each panel, LTC 4 was added at 30 s. wall cells responding). Because repetitive Ca 2+ oscillations can be supported in the absence of Ca 2+ influx, Ca 2+ entry does not directly contribute to their generation. Rather, in agreement with a previous study [22], it is needed to refill the stores and thus maintain the oscillatory mechanism that is driven by the intracellular milieu. At lower agonist concentrations, the pattern of Ca 2+ oscillations was indistinguishable between cells stimulated in 2 mm external Ca 2+,0Ca 2+,and0Ca 2+ together with La 3+ (Figures 1D 1F compares these responses in cells exposed to a low dose of LTC 4 ), although responses in 0 Ca 2+ tended to run down slightly more quickly. We analyzed the amplitude and frequency of the Ca 2+ oscillations for three different LTC 4 concentrations (Figure 2, each point represents >200 cells). The amplitude and frequency of the Ca 2+ oscillations to LTC 4 were generally similar when evoked in either 2 mm Ca 2+ or 0 Ca 2+ and La 3+, whereas responses in 0 Ca 2+ were smaller and shorter lasting. If distinct spatial Ca 2+ gradients hidden within similar global Ca 2+ oscillations encode important information, then one would predict these spatial profiles to differ in their ability to activate Ca 2+ -dependent responses. To test this, we compared the ability of the Ca 2+ oscillations to activate expression of the gene c-fos because (1) it is an important component of the transcription factor complex AP1 that regulates expression of chemokines in immune cells [23] and (2) Ca 2+ influx through CRAC channels is a strong activator of c-fos transcription and translation [24]. Stimulation with LTC 4 for 12 min in the presence of external Ca 2+ consistently activated c-fos expression over a range of agonist concentrations (Figure 3A). Strikingly, stimulation for the same time in the absence of Ca 2+ failed to evoke any c-fos expression even when La 3+ was present to maintain the Ca 2+ oscillations (Figure 3A). This was the case over the range of LTC 4 concentrations tested (Figures 3B 3D). For identical amplitudes and frequencies, only Ca 2+ oscillations in the presence of Ca 2+ influx drove c-fos expression. Moreover, stimulation with 160 nm LTC 4 in Ca 2+ -free solution evoked 7 6 1Ca 2+ oscillations in the 12 min period, whereas 80 nm LTC 4 in 2 mm Ca 2+ generated only oscillations. Despite this, gene expression occurred only after challenge with LTC 4 in the presence of external Ca 2+. The spatial Ca 2+ gradient that arises from local Ca 2+ influx through CRAC channels therefore has two important consequences: (1) it enables lower concentrations of an agonist to evoke a cellular response and (2) fewer Ca 2+ oscillations are needed to activate gene expression, thus increasing the efficacy of excitation-transcription coupling.

3 Calcium Oscillations and Gene Expression 855 Figure 2. The Amplitude and Frequency of Ca 2+ Oscillations to Three Different LTC 4 Concentrations Are Compared (A C) The amplitude of each Ca 2+ oscillation against the oscillation number is plotted. The first Ca 2+ oscillation is presented as peak 0, the second as peak 1, and so on. (A) Response to 160 nm LTC 4 for the three different conditions shown. (B and C) Response to 120 nm and 80 nm LTC 4, respectively. (D F) The number of Ca 2+ oscillations over an 800 s time frame is plotted. Oscillations were binned in 100 s intervals. (D) Response to 160 nm LTC 4. (E and F) Response to 120 nm and 80 nm LTC 4, respectively. Data are presented as mean 6 SEM. Stimulation of cysteinyl leukotriene type I receptors with LTC 4 activates CRAC channels [20] and a low concentration of Gd 3+ (which is considered specific for CRAC channels) accelerated the decline of Ca 2+ oscillations in 2 mm external Ca 2+ (Figure S1). To confirm that Ca 2+ entry through CRAC channels drives gene expression, we blocked the channels with the Synta compound and examined the effects on c-fos induction. As we have shown recently [25], the Synta compound is an effective inhibitor of CRAC channels. Stimulation with 160 nm LTC 4 in Ca 2+ -free solution evoked a series of Ca 2+ oscillations that ran down after w7 min. Readmission of external Ca 2+ resulted in a cytoplasmic Ca 2+ rise resulting from Ca 2+ influx through the open CRAC channels (Figure 4A). Pretreatment with 10 mm Synta compound substantially reduced this Ca 2+ signal. Importantly, c-fos expression in response to LTC 4 (applied in the presence of 2 mm external Ca 2+ ) was prevented by the Synta compound (Figure 4B). Identical results were obtained with a lower concentration of LTC 4 (80 nm). The finding that the amplitude and frequency of Ca 2+ oscillations to LTC 4 were similar in 2 mm Ca 2+ compared with 0 Ca 2+ and La 3+ suggests that Ca 2+ entry and removal mechanisms are in balance. CRAC channels cluster in punctate-like structures upon store depletion [26], but it is not known whether the plasma membrane Ca 2+ pumps or sarco-endoplasmic reticulum Ca 2+ ATPase pumps in peripheral endoplasmic reticulum colocalize here too. Nevertheless, it is likely that Ca 2+ removal pathways are positioned such that they prevent a significant global Ca 2+ increase after Ca 2+ entry, while impacting little on the local subplasmalemmal Ca 2+ rise in the vicinity of open CRAC channels. Clustering of removal mechanisms might be one effective way to ensure that local Ca 2+ entry is impeded from eliciting a more global Ca 2+ rise [27]. We considered the possibility that La 3+ might enter cells and interfere with c-fos expression, thus complicating interpretation of the experiments in Figure 3 where gene expression is measured in 0 Ca 2+ plus La 3+. Two arguments can be raised against this. First, we have previously shown that La 3+ does not enter cells to any detectable level over the time course of the experiments described here [18]. Second, we triggered Ca 2+ influx independent of store-operated Ca 2+ entry to see whether La 3+ inhibited the subsequent activation of gene expression. We used the Ca 2+ ionophore ionomycin, which increases Ca 2+ entry directly through its ionophore properties

4 Current Biology Vol 19 No Figure 3. Gene Expression Occurs Only when Ca 2+ Entry Accompanies the Oscillatory Signal (A) c-fos expression is compared in cells exposed to different concentrations of LTC 4 in 2 mm Ca 2+,0Ca 2+ and La 3+,or0Ca 2+ alone. (B D) The extent of c-fos expression (normalized to control levels in nonstimulated cells) is compared for the different conditions after stimulation in 80 nm (B), 120 nm (C), and 160 nm LTC 4 (D). Over the concentration range tested, gene expression occurred only when accompanied by Ca 2+ influx. Data are presented as mean 6 SEM. *p < 0.01; **p < and indirectly through store depletion and subsequent activation of CRAC channels [28]. Because 1 mm La 3+ completely abolishes store-operated Ca 2+ entry (Figure 4C), we were able to isolate the Ca 2+ influx because of direct ionophore transport by pretreating cells with this trivalent cation. The Ca 2+ response to ionomycin was slightly reduced by 1 mm La 3+ (Figure 4D) but c-fos expression was unaffected (Figure 4E). Hence La 3+ does not interfere with Ca 2+ -induced gene expression. Are global Ca 2+ oscillations necessary for driving gene expression or is local Ca 2+ influx sufficient? To distinguish between these possibilities, we increased cytoplasmic Ca 2+ buffering by loading cells with the slow Ca 2+ chelator EGTA. EGTA reduces the bulk Ca 2+ rise but is too slow to interfere with the amplitude or extent of local Ca 2+ entry [11, 12]. As we have shown previously, preincubating cells with EGTA- AM substantially reduced the Ca 2+ rise that occurred after store-operated Ca 2+ entry [18]. EGTA also abolished the global Ca 2+ oscillations evoked by LTC 4 (63 cells) but had no inhibitory effect on c-fos expression in response to the same concentration of agonist (Figure 4F). We considered the possibility that the c-fos expression response might have already saturated after stimulation with thapsigargin in 2 mm Ca 2+.If so, this could mask a potential difference in c-fos expression when compared with that induced after loading the cytoplasm with EGTA. However, a modest fall in the rate of Ca 2+ entry produced a reduction in the extent of c-fos expression, arguing against saturation of the response. Readmission of 0.25 mm and 0.5 mm external Ca 2+ to cells pretreated with thapsigargin in Ca 2+ -free solution elicited a slower rate of Ca 2+ influx (22.8% 6 6.7% and 48.9% % of that seen in 2 mm Ca 2+ [18]). c-fos expression was likewise reduced to 38% 6 3.8% and 64% 6 5.1% (p < 0.01 in both cases) that seen in 2 mm Ca 2+. The Ca 2+ chelator BAPTA is fast enough to intercept incoming Ca 2+ and thus reduce the amplitude and spatial extent of local Ca 2+ entry [11]. Loading the cytoplasm with BAPTA suppressed gene expression induced by LTC 4 (Figure 4F). Because BAPTA is able to restrict Ca 2+ influx to within a few nanometers of the channel pore, these results demonstrate that local Ca 2+ influx drives gene expression and that global Ca 2+ oscillations are not essential to this process. Our findings provide new insight into how cells decode the universal oscillatory Ca 2+ signal. Rather than responding to the amplitude or frequency of the Ca 2+ oscillations, we have discovered that the spatial profile of the oscillatory Ca 2+ signal is the key trigger for gene expression. Abolition of global Ca 2+ oscillations by increasing cytoplasmic Ca 2+ buffering had no inhibitory effect on gene expression. Instead, local Ca 2+ influx through CRAC channels was sufficient to drive c-fos transcription. The digital nature of the Ca 2+ oscillations is therefore not the fundamental signaling unit used to drive excitation-transcription coupling at low levels of stimulus intensity, at least in the cell type we have used.

5 Calcium Oscillations and Gene Expression 857 Figure 4. Gene Expression Can Occur in the Absence of Ca 2+ Oscillations (A) Cells were stimulated with LTC 4 (160 nm) in Ca 2+ -free solution and once oscillations had stopped, 2 mm Ca 2+ was applied. Store-operated Ca 2+ entry occurred (filled circles, mean of >80 cells) and this was blocked by pretreatment with the Synta compound (10 mm, applied 5 min before Ca 2+ was readmitted; >50 cells). Time 0 on the graph represents 60 s after cells had stopped oscillating. (B) The Synta compound suppresses gene expression induced by LTC 4. Top: Gel shows c-fos induction by 160 nm LTC 4 and this is blocked by the Synta compound. Lower panel summarizes aggregate data from three such experiments. (C) 1 mm La 3+ abolishes store-operated Ca 2+ influx. Thapsigargin (2 mm) was applied in 0 Ca 2+ external solution to deplete the stores and then 2 mm external Ca 2+ was readmitted as indicated. (D) The Ca 2+ signal evoked by 10 mm ionomycin was only partially reduced by La 3+. (E) c-fos expression was induced by ionomycin and this was unaffected by La 3+. (F) Loading cells with EGTA, which prevented the oscillatory response, had no effect on gene expression to LTC 4 whereas loading with BAPTA suppressed c-fos induction. Control refers to LTC 4 in 2 mm Ca 2+. Data are presented as mean 6 SEM. ***p < Although the importance of CRAC channels to physiological responses has been appreciated, it was thought that the role of the channels was simply to provide a top-up of Ca 2+ stores, remedying the small loss of Ca 2+ from the cell after each Ca 2+ oscillation. Our data clearly reveal that the global Ca 2+ rise associated with Ca 2+ oscillations does not provide the Ca 2+ that activates gene expression; rather, it is the spatially localized Ca 2+ signal that arises from open CRAC channels that triggers the nuclear response. Hence CRAC channels do not simply support the signaling function of Ca 2+ oscillations. On the contrary, it appears that the function of Ca 2+ oscillations is to activate CRAC channels, which provides the important and physiological Ca 2+ signal, at least for c-fos gene regulation. The frequency of oscillatory Ca 2+ signals can recruit different transcription factors to the nucleus [5], alter mitochondrial ATP production [6], and drive exocytosis [10]. Our results now show that important information is contained in the spatial signature of the Ca 2+ oscillation. In neurons, spatially restricted Ca 2+ signals at the cell periphery can influence nuclear events: Ca 2+ microdomains arising from L-type Ca 2+ channels [15, 29] or NMDA receptors [30] activate gene expression independent of the bulk Ca 2+ rise. Although these pioneering studies revealed the importance of local Ca 2+ signals in driving excitation-transcription coupling, the stimuli used were mm K + pulses for tens of seconds or bursts of action potentials for 5 min. In our experiments, we have used a physiologically relevant trigger, namely a low dose of an agonist. Spatially restricted local Ca 2+ entry can therefore trigger transcription after a stimulus of physiological intensity. Moreover, it is this spatial signature within a global Ca 2+ oscillation, rather than the oscillation itself, that triggers c-fos gene expression. Such spatial decoding greatly increases the versatility of Ca 2+ as a second messenger, enhances the efficacy of Ca 2+ in promoting gene expression, and provides a novel means for generating specific Ca 2+ -dependent cellular responses. Experimental Procedures Cell Culture The mast cell line RBL-1 was bought from ATCC and cultured as previously described [25]. Single-Cell Ca 2+ Imaging RBL-1 cells were loaded with fura 5F as described [18]. Cells were excited at 356 and 380 nm (20 ms exposures) at 0.5 Hz and the ratio (R) calculated.

6 Current Biology Vol 19 No Standard external solution contained (in mm): NaCl 145, KCl 4.5, CaCl 2 2, MgCl 2 1, HEPES 10, D-glucose 10 (ph 7.2 with NaOH). In 0 Ca 2+ solution, external Ca 2+ was removed and 0.1 mm EGTA (ph 7.2 with NaOH) was added. 0 Ca 2+ +La 3+ was identical to 0 Ca 2+ but 1 mm La 3+ was added (and the ph returned to 7.2 with NaOH because EGTA can bind La 3+ ). LTC 4 was purchased from Axxora. EGTA and BAPTA Loading Cytoplasmic loading with EGTA or BAPTA was carried out by preincubating cells with EGTA-AM or BAPTA-AM (25 mm) for 40 min at w24 C in the dark together with fura 5F (2 mm). Cells were then washed three times and allowed to recover for a further 15 min in the dark. c-fos Expression RT-PCR was carried out as described previously [24]. Supplemental Data Supplemental Data include one figure and can be found with this article online at (09) Acknowledgments This work was supported by an MRC programme grant to A.B.P. J.D.C. held a Christopher Welch Scholarship. S.-W.N. holds a BBSRC-GlaxoSmithKline (GSK) studentship. We thank GSK for the synthesis and provision of the Synta compound. Received: January 20, 2009 Revised: March 19, 2009 Accepted: March 19, 2009 Published online: April 16, 2009 References 1. Berridge, M.J., Bootman, M.D., and Roderick, H.L. (2003). Calcium signalling: Dynamics, homeostasis and remodelling. Nat. Rev. Mol. Cell Biol. 4, Clapham, D.E. (2008). Calcium signaling. Cell 131, Petersen, O.H. (2005). Calcium signalling and calcium-activated ion channels in exocrine acinar cells. Cell Calcium 38, Dolmetsch, R.E., Lewis, R.S., Goodnow, C.C., and Healy, J.I. (1997). Differential activation of transcription factors induced by calcium response amplitude and duration. Nature 386, Dolmetsch, R.E., Xu, K., and Lewis, R.S. (1998). Calcium oscillations increase the efficiency and specificity of gene expression. Nature 392, Hajnoczky, G., Robb-Gaspers, L.D., Seitz, M.B., and Thomas, A.P. (1995). Decoding of cytosolic calcium oscillations in the mitochondria. Cell 82, Prince, W.T., and Berridge, M.J. (1973). The role of calcium in the action of 5-hydroxytryptamine and cyclic AMP on salivary glands. J. Exp. Biol. 58, Berridge, M.J. (1990). Calcium oscillations. J. Biol. Chem. 265, Thomas, A.P., Bird, G.S., Hajnoczky, G., Robb-Gaspers, L.D., and Putney, J.W.J. (1996). Spatial and temporal aspects of cellular calcium signalling. FASEB J. 10, Tse, A., Tse, F.W., Almers, W., and Hille, B. (1993). Rhythmic exocytosis stimulated by GnRH-induced caclium oscillations in rat gonadotropes. Science 260, Neher, E. (1998). Vesicle pools and Ca 2+ microdomains: New tools for understanding their roles in neurotransmitter release. Neuron 20, Parekh, A.B. (2008). Ca 2+ microdomains near plasma membrane Ca 2+ channels: Impact on cell function. J. Physiol. 586, Berkefeld, H., Sailer, C.A., Bildl, W., Rhode, V., Thumfart, J.O., Eble, S., Klugbauer, N., Reisinger, E., Bischofberger, J., Oliver, D., et al. (2006). BKCa-CaV channel complexes mediate rapid and localized calciumactivated potassium signalling. Science 314, Deisseroth, K., Bito, H., and Tsien, R.W. (1996). Signaling from synapse to nucleus: postsynaptic CREB phosphorylation during multiple forms of hippocampal synaptic plasticity. Neuron 16, Dolmetsch, R.E., Pajvani, V., Fife, K., Spotts, J.M., and Greenberg, M.E. (2001). Signaling to the nucleus by an L-type calcium channel-calmodulin complex through the MAP kinase pathway. Science 294, Bautista, D.M., and Lewis, R.S. (2004). Modulation of plasma membrane calcium-atpase activity by local calcium microdomains near CRAC channels in human T cells. J. Physiol. 556, Willoughby, D., and Cooper, D.M.F. (2007). Organization and Ca 2+ regulation of adenylyl cyclases in camp microdomains. Physiol. Rev. 87, Chang, W.C., Di Capite, J., Singaravelu, K., Nelson, C., Halse, V., and Parekh, A.B. (2008). Local calcium influx through calcium releaseactivated calcium (CRAC) channels stimulates production of an intracellular messenger and an intercellular pro-inflammatory signal. J. Biol. Chem. 283, Rizzuto, R., and Pozzan, T. (2006). Microdomains of intracellular calcium: Molecular determinants and functional consequences. Physiol. Rev. 86, DiCapite, J.L., Shirley, A., Nelson, C., Bates, G., and Parekh, A.B. (2009). Intercellular calcium wave propagation involving positive feedback between CRAC channels and cysteinyl leukotrienes. FASEB J. 23, Moreau, B., Straube, S., Fisher, R.J., Putney, J.W., and Parekh, A.B. (2005). Ca 2+ -calmodulin-dependent facilitation and Ca 2+ inactivation of Ca 2+ release-activated Ca 2+ channels. J. Biol. Chem. 280, Bird, G.S., and Putney, J.W.J. (2005). Capacitative calcium entry supports calcium oscillations in human embryonic kidney cells. J. Physiol. 562, Lee, Y.-N., Tuckerman, J., Nechushtan, H., Schutz, G., Razin, E., and Angel, P. (2004). c-fos as a regulator of degranulation and cytokine production in FCeRI-activated mast cells. J. Immunol. 173, Chang, W.-C., Nelson, C., and Parekh, A.B. (2006). Ca 2+ influx through CRAC channels activates cytosolic phospholipase A2, leukotriene C4 secretion and expression of c-fos through ERK-dependent and independent pathways in mast cells. FASEB J. 20, Ng, S.-W., DiCapite, J.L., Singaravelu, K., and Parekh, A.B. (2008). Sustained activation of the tyrosine kinase Syk by antigen in mast cells requires local Ca 2+ influx through Ca 2+ release-activated Ca 2+ channels. J. Biol. Chem. 283, Lewis, R.S. (2007). The molecular choreography of a store-operated calcium channel. Nature 446, Petersen, O.H., and Tepikin, A. (2008). Polarized calcium signalling in exocrine gland cells. Annu. Rev. Physiol. 70, Vig, M., DeHaven, W.I., Bird, G.S., Billingsley, J.M., Wang, H.J., Rao, P.E., Hutchings, A.B., Jouvin, M.H., Putney, J.W., and Kinet, J.P. (2008). Defective mast cell effector functions in mice lacking the CRACM1 pore subunit of store-operated calcium release-activated calcium channels. Nat. Immunol. 9, Deisseroth, K., Heist, E.K., and Tsien, R.W. (1998). Translocation of calmodulin to the nucleus supports CREB phosphorylation in hippocampal neurons. Nature 392, Hardingham, G.E., Arnold, F.J.L., and Bading, H. (2001). A calcium microdomain near NMDA receptors: On switch for ERK-dependent synapse-to-nucleus communication. Nat. Neurosci. 4,